The term "projective test" has become overloaded with connotations that automatically, with little regard for the facts, elicit positive or nega tive responses. This irrationality has already been amusingly documented by Levy & Orr's (76) observation that psychologists in clinical versus aca demic settings are inclined to obtain opposing results when evaluating Ror schach test hypotheses. In view of accumulating work on experimenter bias (62), it should come as no surprise that strong prejudices about projective tests have intruded in such a fashion into experimental results. One could argue convincingly that the term "projective test" should be abolished not only because of the passion it arouses but also because it has seriously cracked under the strain of trying to hold together an impossible conglomer ate of meanings. Even in the early days of projective test development there was, as Zubin, Eron & Schumer (150) and Murstein (94) have pointed out, confusion about what made a response "projective." There has been continuing uneasiness about how to include under one rubric measures which range from perceptual to motor, responses that vary from sponta neous to those elicited by direct questioning, interpretative frameworks which argue opposingly for the paramount importance of the form versus the content of the response, and orientations which see validity only in pat terned interpretation of responses as contrasted to views that the future of proj ective tests depends upon isolating simple unidimensional variables. The diversity is so great that it is not an exaggeration to say that some "projective tests" resemble the MMPI more than they do other "projective tests." For example, the procedure of compiling, without thought as to their logic, a list of "signs" which will distinguish the figure drawings of schizophrenic and normal subjects is much closer to the logic and design of the MMPI than it is to the Rorschach rationale for determining the experi ence balance.Looking back, one fi nds that a single projective test like the Rorschach has served such diverse uses as the following: (a ) an impressionistic clini cal instrument in which only the interpreter's uprightness and verbal fluency limit his deductions; (b) a source of carefully defi ned samples of imagery indicating how much an individual is caught up with one motive or affect state (e.g., anxiety, hostility); (c) a means for detecting the im pact of subliminal stimuli (130); (d) a context for observing every possible kind of response deviance; (e) a procedure for assaying harmony between married couples (80); (f) a measure of intelligence; (g) a means for determining the character of the body image boundary (45); (h) an index 1 The survey of the literature was concluded in March 1966. 165 104 Annu. Rev. Psychol. 1967.18:165-190. Downloaded from www.annualreviews.org Access provided by McMaster University on 02/05/15. For personal use only. Quick links to online content Further ANNUAL REVIEWS 166 FISHER of commonality of fantasy among various cultures (...